2161. How Good is Ultrasound FNA for Neck Lymph Nodes? Single-Institution Review of Diagnostic Adequacy and Associated Features
Authors * Denotes Presenting Author
  1. Hailey Choi; University of California San Francisco
  2. Joshua Chen *; University of California San Francisco
  3. Chienying Liu; University of California San Francisco
  4. Elham Khanafshar; University of California San Francisco
  5. Tara Morgan; University of California San Francisco
Small cervical lymph nodes are frequently found on imaging, raising alarm for patients with known or treated malignancy, such as thyroid cancer. In many instances, ultrasound-guided fine needle aspiration (FNA) is necessary for definite diagnosis, but FNA can be technically challenging and limited by hypocellular or acellular samples. Our objective was to determine the rate of cytologic and diagnostic adequacy and identify features associated with suboptimal tissue sampling in ultrasound-guided FNA of suspected nodal disease in patients with thyroid cancer.

Materials and Methods:
The pathology database was queried for neck FNA reports in patients with history of thyroid cancer from 2014-2019. Subjects/lesions with restricted chart access, incomplete pre-procedure imaging, non-ultrasound-guided FNA, or FNA of thyroid nodules were excluded. Chart review was conducted for patient age, gender, BMI, prior thyroidectomy, primary cancer diagnosis, subsequent surgical excision and surgical pathology diagnosis. Images were retrospectively reviewed for number of FNA passes, lesion location, size, distance from skin, and presence of cystic components, macrocalcification, echogenic foci, and internal vascularity. Per-lesion analysis was performed with ANOVA and Chi-sq tests in R.

Initial query yielded 552 lesions in 343 subjects. Following exclusion, 379 lesions in 210 subjects were included. Preliminary analysis of 185 lesions in 124 subjects is presented for this abstract. Patient age averaged 47.6 years (range 19-90) and BMI 27.4 (range 17.1-47.3); 73.5% were female and 26.5% male. Majority (96.8%) had PTC, and 67.6% had a prior thyroidectomy. Pathology results were categorized into: cellular and diagnostic (157 lesions), hypo/acellular and non-diagnostic (12), hypocellular but diagnostic (9), and hypo/acellular but diagnostic based on thyroglobulin washout (7). 15.1% of lesions were hypo/acellular, and 6.48% were non-diagnostic on FNA. No patient-related factors showed significant association (p > 0.05). More FNA passes (2.54 vs. 1.95), right-sided lesions (67.9% vs. 45.2%), higher long-axis/short-axis size ratio (2.84 vs. 2.31), and presence of cystic foci (53.6% vs. 31.2%) were significantly associated (p < 0.05). Short-axis size alone did not show significance. Lesions < 5 mm short-axis were 18.4% hypo/acellular, versus 14.0% for larger lesions. 15 (53.6%) of hypo/acellular samples underwent surgical excision. Aside from 4 lesions with non-diagnostic FNA, the remainder of FNA diagnoses were concordant with surgical pathology.

Ultrasound-guided FNA has high diagnostic yield (93.5%) and obtains cellular samples in 84.1% of cases. Even for sub-5 mm lesions in the neck, cytologically adequate samples are obtained (81.6%). Features associated with suboptimal cellularity include: higher long/short-axis ratio and cystic foci, with a right-sided predilection. Identifying these features and expected successful sample rates can better inform next step management decisions in thyroid cancer patients.